Related papers: JEPA-T: Joint-Embedding Predictive Architecture wi…
We present \textsc{Vx2Text}, a framework for text generation from multimodal inputs consisting of video plus text, speech, or audio. In order to leverage transformer networks, which have been shown to be effective at modeling language, each…
Text-to-image synthesis aims to generate natural images conditioned on text descriptions. The main difficulty of this task lies in effectively fusing text information into the image synthesis process. Existing methods usually adaptively…
Diffusion models for Text-to-Image (T2I) conditional generation have recently achieved tremendous success. Yet, aligning these models with user's intentions still involves a laborious trial-and-error process, and this challenging alignment…
Typical methods for text-to-image synthesis seek to design effective generative architecture to model the text-to-image mapping directly. It is fairly arduous due to the cross-modality translation. In this paper we circumvent this problem…
Panorama generation has recently attracted growing interest in the research community, with two core tasks, text-to-panorama and view-to-panorama generation. However, existing methods still face two major challenges: their U-Net-based…
Learning to construct text representations in end-to-end systems can be difficult, as natural languages are highly compositional and task-specific annotated datasets are often limited in size. Methods for directly supervising language…
Large text-to-image models achieved a remarkable leap in the evolution of AI, enabling high-quality and diverse synthesis of images from a given text prompt. However, these models lack the ability to mimic the appearance of subjects in a…
While MLLMs perform well on perceptual tasks, they lack precise multimodal alignment, limiting performance. To address this challenge, we propose Vision Dynamic Embedding-Guided Pretraining (VDEP), a hybrid autoregressive training paradigm…
Diffusion models have revitalized the image generation domain, playing crucial roles in both academic research and artistic expression. With the emergence of new diffusion models, assessing the performance of text-to-image models has become…
The integration of generative artificial intelligence (AI) into architectural design has advanced significantly, enabling the generation of text, images, and 3D models. However, prior AI applications lack support for text-to-parametric…
Self-Supervised Learning (SSL) has shifted from pixel-level reconstruction to latent space prediction, spearheaded by the Joint Embedding Predictive Architecture (JEPA). While effective, standard JEPA models typically rely on a…
Recent advances in image tokenizers, such as VQ-VAE, have enabled text-to-image generation using auto-regressive methods, similar to language modeling. However, these methods have yet to leverage pre-trained language models, despite their…
The goal of a speech-to-image transform is to produce a photo-realistic picture directly from a speech signal. Recently, various studies have focused on this task and have achieved promising performance. However, current speech-to-image…
Invariance-based and generative methods have shown a conspicuous performance for 3D self-supervised representation learning (SSRL). However, the former relies on hand-crafted data augmentations that introduce bias not universally applicable…
Discrete image tokenizers have emerged as a key component of modern vision and multimodal systems, providing a sequential interface for transformer-based architectures. However, most existing approaches remain primarily optimized for…
Customized text-to-image generation, which aims to learn user-specified concepts with a few images, has drawn significant attention recently. However, existing methods usually suffer from overfitting issues and entangle the…
Unified models aim to support both understanding and generation by encoding images into discrete tokens and processing them alongside text within a single autoregressive framework. This unified design offers architectural simplicity and…
Text-to-image generative models have achieved remarkable visual quality but still struggle with compositionality$-$accurately capturing object relationships, attribute bindings, and fine-grained details in prompts. A key limitation is that…
Two competing paradigms exist for self-supervised learning of data representations. Joint Embedding Predictive Architecture (JEPA) is a class of architectures in which semantically similar inputs are encoded into representations that are…
Continual post-training adapts a single text-to-image diffusion model to learn new tasks without incurring the cost of separate models, but naive post-training causes forgetting of pretrained knowledge and undermines zero-shot…